The accurate calculation and delivery of dose in radiotherapy patients is critical in order to maximize dose to the target volume and minimize dose to surrounding normal tissue. Dose calculation errors can lead to normal tissue complications or underdosage of the target volume leading to local recurrence of disease. The computation of dose from electron therapy beams is particularly difficult, and even the most advanced algorithm can be in error by 12% with simple tissue inhomogeneities. The clinical criteria for acceptable dose calculation accuracy is 3% in all anatomic locations. This project has the ultimate aim of producing an electron dose computation algorithm whose accuracy lies within the 3% criteria for patient treatment. Recently two new theoretical models which more accurately describe electron penetration in tissue have been developed by our group. These models are called the Restricted Scattering Model and the Electron Loss Model and they have distinct advantages over the theory most commonly used for electron beam dose calculations, Fermi-Eyges theory. A major advantage of the Electron Loss Model over Fermi-Eyges theory is that measured electron dose distributions are not required to supplement the theory of electron penetration since the model is inherently able to predict the variation of dose with depth in tissue at therapeutic beam energies. Further development and numerical implementation of dose algorithms based on these new models are required. This project proposes to develop, improve, implement and verify these new algorithms in order to reduce electron dose computation errors to an acceptable level of +3%. Three main areas of interest are identified; (1) the further development and implementation of the two new models of electron penetration in patients, (2) verification with measured dose data and Monte Carlo simulations of the predictions of the new electron penetration models, (3) the modeling of electrons scattering in the air drift space to account for beam geometry, energy and collimation in scatter-foil accelerators.